Smart environments should be able to identify the users so that they can deliver personalized services. For example, a smart TV can recommend programs based on the identity of the user; a smart store can recognize its regular customers and provide more personalized services; a smart home can identify the family members and remind them of important messages.
In order to achieve the smart environment, a system which supports and realizes user authentication and registration is very critical. A simple and common method is that a user inputs username and password by keyboard or another input device. However, this manner requires direct user intervention. Another authentication device performs authentication by using both physical characteristics such as a fingerprint and behavioural characteristics such as “motion” implemented by sliding a finger on the device (or sliding the device on a finger).
Face recognition is a substitution method for user authentication and registration since it is unobtrusive and allows the user to remain relatively passive. In a face recognition system, a face sample database is necessary. If there is no a pre-stored face database, face recognition cannot be conducted. In fact, in a public environment (such as in museum), it is hard to establish a public face database. In domestic environment, in the phase of initialization, there is also no face database.
Another method is to allow a user to take advantage of an RFID card or another wireless device which can indicate user identity. In a smart TV scenario, an RFID card can help complement user registration automatically. However, if there are several persons in front of the RFID reader, it is hard to distinguish who among the several users holds an RFID card and what is the relationship between the detected cards and the persons present by wireless signal analysis only.
Therefore, it is desirable to come up with a solution that can implement user authentication and registration effectively.